Google has released a new artificial intelligence chatbot called Bard, which will compete with OpenAI’s ChatGPT. The company has opted for a quiet release, with select insiders receiving demos and a “Join waitlist” button for others to gauge their interest.

The move comes after the success of ChatGPT. This release prompted a “code red” within Google, as it sought to release a similar product. However, the initial announcement of Bard proved faulty when the AI made an incorrect factual assertion during a moving graphic intended to showcase Google’s AI capabilities. The incident caused a drop in the parent company’s shares by $100 billion.

AI hallucination is a term that refers to when an AI generates false or misleading information that appears to be accurate. This can happen when an AI system is fed incorrect or biased data during its training, causing it to produce inaccurate responses. AI hallucinations can also occur when an AI system generates responses that are based on incomplete or outdated information, or when it is programmed to generate responses that are intentionally misleading or deceptive.

The Bard homepage states that “Bard is experimental” and warns that some responses may be inaccurate or offensive. When using the AI chatbot, users may find the experience similar but different from ChatGPT and Bing.

In early tests, Bard was found to be creative and could generate ideas. However, its writing skill wasn’t quite at the level of its namesake, and some of the AI’s responses were repetitive.

It’s worth noting that Microsoft Bing’s AI chatbot also had its own misbehavior, however, Bing is receiving more traffic than ever before. As with all AI chatbots, users should approach them with caution and understand that there may be inaccuracies or offensive responses.

As a fail-safe, Google implemented a multiple-answer feature to its AI. The Bard can give multiple answer options for those seeking answers to basic questions from the AI. 

To get access to Bard, users can click on the “Join waitlist” button on the Bard homepage. The site will notify users if their account is eligible and ask them if they want to opt-in for Bard news updates. Once on the waitlist, users will receive an email with the subject line “It’s your turn to try Bard” when it’s their turn to test the AI chatbot.

The waitlist is expected to be long, as Google is only releasing Bard to a limited test group. Google’s move to release Bard is part of a larger push by the company to develop AI chatbots that can converse with humans in a natural and engaging way.

While AI chatbots have come a long way, there is still much work to be done before they can fully replace human interaction.